Research On Environmental Control System Of Piggery Based On Multi-parameter Fusion

With the development of pig breeding from individual to large-scale intensive piggery,the environmental problem of piggery has been one of the most important factors affecting pig breeding.The piggery environmental controller also begins to develop from a single control to an integrated multi-dimensional control.At the same time,due to the development and application of artificial intelligence,the frontier technologies such as fuzzy set,neural network and data fusion are gradually applied in the piggery environmental control,making the control technology gradually intelligent.Based on the development of breeding pig industry both at home and abroad in this paper,a multi-parameter fusion of piggery environmental control system,realize the piggery environmental parameter acquisition and storage,LCD man-machine interface real-time view piggery scene environment parameter information,through the Internet of things platform remote access piggery live environment,the introduction of data fusion technology of piggery environmental status identification,Then control the environment control equipment to start and stop to achieve the pigsty environmental state of the appropriate stable state.The data fusion technology based on D-S evidence theory was studied.Firstly,the state frame of piggery environment identification was established based on the range of environmental parameters specified by national standards.Then the normal membership function of the fuzzy set is used to extract the feature information from the evidence source and generate the basic probability distribution function of the elements in the recognition frame.In view of the conflicts existing in the fusion of D-S evidence theories,the distance between K-L evidence is calculated to allocate the weight of evidence fusion,reduce the fusion conflict and ensure the consistency of evidence fusion.By using the improved D-S evidence combination framework,the conflicting parts are equally distributed to each recognition focal element in the fusion process,and the distributed fusion mechanism is adopted to fuse the evidence one by one,and finally the fusion results are obtained.The hardware part is composed of the main control chip STM32F429IGT6 and other modules,and the software part is based on the Free RTOS real-time operating system to start the task scheduling.Hardware modules mainly include data storage module,sensor module,Internet of Things module,LCD module,output control module and so on.At the software level,data acquisition/storage tasks,LCD display tasks,Internet of Things uploading tasks,data fusion tasks,control output tasks,timing tasks are realized in the real-time system,and the overall realization of data acquisition-storage-display-upload-fusion control functions.The equipment was installed in a pigsty test in Hubei Province to verify the pigsty control effect.The results showed that the fluctuation rate of temperature,humidity,ammonia concentration,hydrogen sulfide concentration and carbon dioxide concentration in the pigsty was small,and the pigsty was within the suitable environment range.The data fusion output identifies the piggery environment state close to the real piggery environment state.The results show that the piggery environmental state identification based on data fusion meets the expected requirements,and the piggery environmental control system has a good application effect.